While ML-based decisions are made at the inspection step, cloud-based ML algorithms derive the edge compute algorithms.<\/p>\n
\u201cUsing AI and machine learning in semiconductor inspection and metrology is no longer a question of whether they should be used, but how it should help,\u201d said Charlie Zhu, senior director of product engineering for advanced technology solutions at Nordson Test & Inspection<\/a>. \u201cSimilar to other industry players, we are pushing more data into the cloud. Definitely, there\u2019s a tradeoff between cloud and edge compute. Inspection and the measurement will continue to be performed with edge computing, especially for our products, which provide in-line 100% inspection. Computing at the edge is still faster. Model training will be preferable on the cloud because of the GPU computational power needed to train a model. But once trained, the computation power for inferencing is less demanding.\u201d<\/p>\n The balance of compute on the cloud versus the edge varies with respect to measurement and decision goals.<\/p>\n \u201cFrom my experience, equipment vendors have always wanted data to be made available on the cloud to help them troubleshoot equipment recipes, etc.,\u201d said Aftkhar Aslam, CEO of yieldWerx<\/a>. \u201cIDMs have indicated the need to have data available on the cloud where they can do cross-fabrication correlation and root-cause analysis. A recommendation would be a hybrid approach, where key data for a specific problem \u2014 early technology introduction, NPI lifecycle stage products where there is huge overlap, or correlation to process yields \u2014 makes sense to keep data on the cloud versus the edge.\u201d<\/p>\n Others agree that a hybrid computing architecture approach fits most computing needs, which may vary depending upon the amount of data and the application.<\/p>\n \u201cThere is no one-size-fits-all approach,\u201d said Steve Zamek, director of technical product management at PDF Solutions<\/a>. \u201cA hybrid architecture with an enterprise-wide platform that allows deploying models to the edge may offer the best of all worlds. These considerations are not unique to AL\/ML models. Many of our customers used similar approach to deploy rule-based models years ago. However, model sizes have been increasing, and training some of the larger models is only possible in a scalable and centralized infrastructure, i.e., the cloud.\u201d<\/p>\n Table 1: Pros and cons of different deployment options. Green is good, yellow is okay, red is poor. Source: PDF Solutions<\/p>\n In the cloud There is also a growing trend to combine inspection and metrology data with data collected upstream and downstream, then use it to unveil subtle defects. This pushes more computing to the cloud, illustrating the need for data infrastructure platforms that draw from multiple data sources. Here, data quality is paramount.<\/p>\n \u201cML-based inspection relies on pre-trained defect models stored in a library to recognize defects. Unlike traditional methods that depend on pattern repetition, ML algorithms analyze features from a diverse set of training images, making them well-suited for inspecting partial dies and wafer edges,\u201d noted Woo Young Han, product marketing director at Onto Innovation<\/a>. \u201cAdditionally, because ML models are trained to recognize specific defect types, defect classification occurs simultaneously with the inspection process, improving efficiency and accuracy.\u201d<\/p>\n Assembling all the necessary images to build advanced ML presents a daunting investment for cost-conscious manufacturing facilities. And the complexity of the data infrastructure only grows with chiplet-based products, which source die from multiple fabs.<\/p>\n \u201cThe biggest hurdle preventing the customer adopting AI nowadays is the upfront cost,\u201d said Nordson\u2019s Zhu. \u201cI\u2019m not talking about the monetary cost, but about the effort to collect all the data. You need a huge amount of data to train these models. Some models require hundreds of thousands, or up to millions of image pieces. We are solving this problem by providing a generic model. We do the heavy lifting by training a model on the data we have available. But not all the model development can be done this way. It depends upon the application. For instance, we found that all the PCBs look similar in terms of component types. There is a limited number of package types (e.g. QFPs, QFNs) used per the IPC standards. We collect all these PCB component image data and train a generic model which can do segmentation on any PCB board.\u201d [1]<\/p>\n Fig. 1: AOI PCB segmentation using AI to segment\/label features in AOI image. Source: Nordson Test and Inspection<\/p>\n Combining inspection image data with electrical test data has become standard practice in model building. This additional information provides model input to discern nuisance from impactful defects.<\/p>\n \u201cTake a simple task of image classification,\u201d said PDF\u2019s Zamek. \u201cFor model training, one may use electrical test as the \u2018ground-truth\u2019 of whether the defect is a killer or a nuisance. To do so, one needs to gather electrical test data from a variety of steps, including wafer sort, package-level test, burn-in, and so on. And these data have to be gathered from multiple sites, ideally in a cloud for ease of use. Training requires large volume of images to cover different process technologies, inspection methods and equipment, inspection recipes, and many more. This necessitates access to scalable compute resources, driving toward cloud solutions again.\u201d<\/p>\n Once built, the model can be applied at the point of inspection on an edge computer. However, continuous improvement is required. The model needs an occasional update based on data from multiple inspection\/metrology tools, often from multiple manufacturing facilities. That data is fed back to the cloud, the model is modified, and then it\u2019s deployed in tools in the field.<\/p>\n Connecting more data \u201cThe key challenge for inline metrology and inspection (in fab and foundry) is that the models trained and deployed on the equipment are limited to the data types available on that equipment, which is quite limited,\u201d observed PDF\u2019s Zamek. \u201cWe\u2019ve been providing a platform that enables us to bring all data from all operations, from all sites, under one roof. And we have been seeing a growing number of use cases building and deploying models to correlate metrology to PCM, inline inspection to yield, and so on.\u201d<\/p>\n Fig. 2: Typical manufacturing data pipelines that feed into the cloud for building models across factories. Source: PDF Solutions<\/p>\n Put simply, combining data from multiple sources within a factory and across factories has proven benefits.<\/p>\n \u201cFundamentally, data analytics approaches have evolved significantly with the onset of AI and machine learning models,\u201d said Melvin Lee Wei Heng, director of field applications for enterprise software at Onto Innovation. \u201cThese models have greatly enhanced traceability, making it a crucial aspect of macro-defect detection and corrective actions. The ability to link and connect information from back-end to front-end processes has enabled factories to implement predictive models at the front end, even before parts arrive at the back-end process. This integration has improved response times and decision-making accuracy, leading to more efficient and effective defect management.\u201d<\/p>\n At the edge \u201cThere\u2019s always a need for rapid decision-making on inspection and metrology data, closing the loop quickly to pinpoint which process step is driving defects, determine if rework is required, and assess the impact on current WIP,\u201d said yieldWerx\u2019s Aslam. \u201cThe concern with relying solely on the cloud is clear \u2014 security, network latency, and potential inaccessibility. If data becomes unavailable, lots and equipment may be forced on hold until access is restored, often at a significant cost.\u201d<\/p>\n Just as test systems have added a computing box alongside the ATE, inspection and metrology equipment vendors now provide a separate and local GPU computing resource.<\/p>\n \u201cTo maintain high throughput, ML-based inspection requires a separate graphics processing unit (GPU), which operates in parallel with conventional inspection techniques,\u201d said Onto\u2019s Han. \u201cThis parallel processing approach ensures that the use of ML does not negatively impact throughput while enhancing defect detection and classification capabilities.\u201d<\/p>\n GPUs have simply become a necessity to support localized decision-making. \u201cWithin our product portfolios, AI has shifted workloads associated with image processing and data extraction to GPUs, enhancing the efficiency and performance of image computers,\u201d said the KLA spokesperson. \u201cThese GPU-based image computer architectures are \u2018edge\u2019 compute systems supporting real-time data processing and the use of AI algorithms that produce instantly accessible data streams for inline monitoring, benefitting semiconductor manufacturers\u2019 time-to-results and yield entitlement.\u201d<\/p>\n Conclusion The trend to combine inspection data with other equipment data throughout the manufacturing process (e.g., electrical test) necessitates a centralized data lake that can use scalable cloud compute resources. The resulting AI\/ML model improves the detection of impactful defects. Subsequently, the inspection system deploys the model at the edge with nearby GPU resources that can feed into a factory\u2019s yield management systems.<\/p>\n The positive impact on yield and quality improvement in manufacturing facilities is unquestioned. \u201cUltimately, ML-driven analysis techniques for metrology and inspection will always vary greatly at the measurement level,\u201d said Sean King, principal product manager at Onto Innovation. \u201cHowever, as process complexity and data volumes skyrocket, using AI and ML to discern patterns and more intelligently analyze results in context becomes more common between the methods. Yield becomes less about atomized optimization of defects and process steps, and more about the holistic \u2018yield space\u2019 of intertwined (and not always clearly correlated) factors.\u201d<\/p>\n Reference<\/p>\n https:\/\/www.electronics.org\/ipc-standards<\/a>, note IPC is now Global Electronics Association<\/p>\n Related Reading <\/p>\n","protected":false},"excerpt":{"rendered":"An explosion in data from inspection images and metrology measurements is creating a confusing set of demands for…\n","protected":false},"author":2,"featured_media":199043,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[21],"tags":[4323,86,56,54,55],"class_list":{"0":"post-199042","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-computing","8":"tag-computing","9":"tag-technology","10":"tag-uk","11":"tag-united-kingdom","12":"tag-unitedkingdom"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/199042","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/comments?post=199042"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/199042\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media\/199043"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media?parent=199042"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/categories?post=199042"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/tags?post=199042"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"\"](\"https:\/\/www.newsbeep.com\/uk\/wp-content\/uploads\/2025\/10\/Screenshot-2025-10-13-at-4.17.58-PM.png\")
<\/p>\n
For difficult image analysis challenges, defect detectability significantly improves with advanced ML algorithms. ML model development requires hundreds of thousands of relevant images, and this is where cloud computing really shines, offering efficient GPU-based computing for processing massive volumes of data.<\/p>\n![\"\"](\"https:\/\/www.newsbeep.com\/uk\/wp-content\/uploads\/2025\/10\/Screenshot-2025-10-13-at-4.18.16-PM.png\")
<\/p>\n
With the ability to draw data from multiple sources, engineering teams can develop advanced ML models to uncover relationships between equipment parameters upstream, and image data and electrical test data downstream. This can identify abnormalities and speed root-cause analysis within a factory.<\/p>\n![\"\"](\"https:\/\/www.newsbeep.com\/uk\/wp-content\/uploads\/2025\/10\/Screenshot-2025-10-13-at-4.18.34-PM.png\")
<\/p>\n
Models are built on the cloud and applied at the edge. Moving data from inspection\/metrology systems to the cloud to make a decision to execute back at the system is simply impractical. For quick corrective action, inspection and metrology decisions need to be connected with upstream process data as close to real-time as possible.<\/p>\n
Successfully applying AI\/ML for inspection applications requires both cloud and edge computing resources and accumulated images. For model building, the cloud uses at least 100,000 images, and more often a million.<\/p>\n
AI\/ML Challenges In Test and Metrology
<\/a>New tools are changing the game, but it will take time and collaboration for them to achieve their full potential.
Metrology Under Pressure: Detecting Defects in Fine-Pitch Hybrid Bonding
<\/a>Shrinking interconnects expose limitations in traditional inspection methods, forcing new approaches to overlay, surface quality, and defect detection.<\/p>\n